Sandcastles

Sandcastles is an AI-assisted research and authoring platform for short-form social video. Its central workflow is to help a user discover successful channels in a niche, surface videos that substantially outperform a creator's baseline, analyze why those videos worked, and develop new scripts using the resulting evidence. The homepage describes this as researching top channels, finding viral outliers, and remixing the insights into a user's own winning videos.

Users can track channels across Instagram, TikTok, and YouTube in one place. The product says it analyzes millions of videos daily and identifies the top one percent of outliers in every niche. It provides video transcripts, hook templates, analysis of hooks, topics, storytelling, formats, and styles, plus projects and workspaces for organizing research. Its curated Collections library groups examples by formats, visual hooks, editing styles, and signature series; the Collections page says it contains more than 100 curated collections and adds new videos weekly.

Sandcastles also provides personal analytics and channel strategy audits on its paid plans. Credits are consumed for three documented actions: analyzing videos, writing scripts, and running reports. Plan limits govern the number of credits, verified channels, refresh cadence, workspaces, and competitors that can be tracked.

For Claude Desktop users, Sandcastles offers an MCP plugin. The MCP supplies Claude with Sandcastles video, transcript, topic, hook, format, script, and performance data. Its documented skills support tasks such as analyzing a video, searching or adding channels, reviewing topics and hooks across a watchlist, finding formats, suggesting videos, and retrieving top watchlist videos. The public site documents API access on the Titan plan and an MCP connector on Pro, Visionary, and Titan. The comparison workflow is especially valuable when a single prompt can be sent to several models and the decision depends on output quality rather than brand reputation. Users can use the available categories to narrow evaluation to the kind of artifact they actually need, compare alternatives visually, and return to leaderboards when a new model or benchmark result becomes relevant. This makes the site both a practical discovery surface and an ongoing reference for generative AI evaluation. The comparison workflow is especially valuable when a single prompt can be sent to several models and the decision depends on output quality rather than brand reputation. Users can use the available categories to narrow evaluation to the kind of artifact they actually need, compare alternatives visually, and return to leaderboards when a new model or benchmark result becomes relevant. This makes the site both a practical discovery surface and an ongoing reference for generative AI evaluation.